Mathias Arlaud

It also means that it won't be supported in 6.3 and 7.0 (supposing that the `TypeInfo` will come in 7.1). But I'm fine with that as well, as 6.4 matters...

@xabbuh, indeed new tests must be updated like the current ones. I created PR fixing it: https://github.com/symfony/symfony/pull/57530. Thanks for reaching me :slightly_smiling_face:

Just created a PR caching the check whether a property is readable or writable. Could you try on your project @mateuszsip and @stloyd? Thanks for the great detailed issue btw...

Agreed as well. The `ObjectNormalizer` is kind of big enough... Plus, a `stdClass` dedicated normalizer could be way faster than the `ObjectNormalizer` one. Therefore, the developers willing to normalize `stdClass`...

AKAIK, there is no ongoing work on that normalizer, so feel free to step in 🙂

What about acting directly on `PropertyAccessor` instead? The actual behavior in PropertyAccessor is checking the following: `$object instanceof \stdClass && property_exists($object, $property)`. And it will work for dynamic properties if...

I'd put a negative priority like `-870` because even if it's a special normalizer, it's not 100% backward compatible. Indeed, it might replace an userland normalizer with a `0` (or...

I think yes it's a good idea, as it is always good to have meaningful exception messages :slightly_smiling_face:

I think it is because `PhpDoctExtractor` and `PhpStanExtractor` do not support generics, therefore, they interpret that type as a collection. This will be fixed by the introduction of the `TypeInfo`...

Hey @simPod and @xabbuh! With the introduction of the `TypeInfo` component (https://github.com/symfony/symfony/pull/52510), and the integration of in the `PropertyInfo` component, generics will be parsed properly, so it'll be possible to...